Oncoprints

Chante Bethell for CCDL 2019

This notebooks creates oncoprints given the relevant metadata and output MAF files from the snv callers strelka2, mutect2, lancet, and vardict. It addresses issue #6 in OpenPBTA.

Output Files

• analyses/oncoprint-landscape/plots/combined_maf_oncoprint.png
• analyses/oncoprint-landscape/plots/strelka2_oncoprint.png
• analyses/oncoprint-landscape/plots/mutect2_oncoprint.png
• analyses/oncoprint-landscape/plots/lancet_oncoprint.png
• analyses/oncoprint-landscape/plots/vardict_oncoprint.png

Usage

This script is intended to be run via the command line from the top directory of the repository as follows:

Rscript -e "rmarkdown::render('analyses/oncoprint-landscape/01-plot-oncoprint.Rmd', 
                              clean = TRUE)"

Set Up

# Install maftools
if (!("maftools" %in% installed.packages())) {
  if (!requireNamespace("BiocManager", quietly = TRUE)) {
    install.packages("BiocManager")
  }
  BiocManager::install("maftools")
}
library(maftools)

# Get `magrittr` pipe
`%>%` <- dplyr::`%>%`

# Define a color vector for plots 
colores = c("Missense_Mutation" = "#35978f", 
        "Nonsense_Mutation" = "#191970",
        "Frame_Shift_Del" = "#56B4E9", 
        "Frame_Shift_Ins" = "#FFBBFF", 
        "Splice_Site" = "#F0E442",
        "Nonstop_Mutation" = "#56B4E9",
        "In_Frame_Del" = "#CAE1FF",
        "In_Frame_Ins" = "#FFE4E1",
        "Multi_Hit" = "#f46d43")

Create function to plot oncoprints

plot_oncoplot <- function(maf_file, filename){
  # Given a maf file and a filename, plot an oncoprint of the variants in the 
  # dataset, save as a png file and display.
  # Args:
  #   maf_file: name or path to a maf file
  #   filename: name to save the png file as 
  # Return:
  #   oncoprint: plot produced using the maftools `oncoplot` function
  
  # Plot and save the oncoprint 
  png(file.path(plots_dir, filename), width = 60, height = 30, units = "cm", res = 300)
  oncoplot(maf_file,
  clinicalFeatures = c(
    "broad_histology",
    "short_histology",
    "reported_gender",
    "tumor_descriptor"
  ),
  logColBar = TRUE,
  sortByAnnotation = TRUE,
  showTumorSampleBarcodes = TRUE,
  removeNonMutated = FALSE,
  annotationFontSize = 0.7,
  SampleNamefontSize = 0.5,
  fontSize = 0.7,
  colors = colores
  )
  dev.off()
  
  # Display saved oncoprint
  knitr::include_graphics(file.path(plots_dir, filename))
}

Directories and Files

# Path to the data obtained via `bash download-data.sh`.
data_dir <- file.path("..", "..", "data")

# Path to output directory for plots produced
plots_dir <- "plots"

if (!dir.exists(plots_dir)) {
  dir.create(plots_dir)
}

Read in data

# Read in metadata
metadata <- readr::read_tsv(file.path(data_dir, "pbta-histologies.tsv"))

# Rename for maftools function
metadata <- metadata %>%
  dplyr::rename("Tumor_Sample_Barcode" = "Kids_First_Biospecimen_ID")

# Read maf files into a list
maf_list <- list(
  strelka2 = read.maf(
    file.path(data_dir, "pbta-snv-strelka2.vep.maf.gz"),
    clinicalData = metadata,
    verbose = FALSE
  ),
  mutect2 = read.maf(
    file.path(data_dir, "pbta-snv-mutect2.vep.maf.gz"),
    clinicalData = metadata,
    verbose = FALSE
  ),
  lancet = read.maf(
    file.path(data_dir, "pbta-snv-lancet.vep.maf.gz"),
    clinicalData = metadata,
    verbose = FALSE
  ),
  vardict = read.maf(
    file.path(data_dir, "pbta-snv-vardict.vep.maf.gz"),
    clinicalData = metadata,
    verbose = FALSE
  ),
  combined = read.maf(
    file.path("data", "pbta-snv-combined.vep.maf.gz"),
    clinicalData = metadata,
    verbose = FALSE
  )
)

Plot, Display, and Save Oncoprints

# Plot oncoprint using combined maf output data and save. This combined data 
# was created using the `cat` function via command line to combine the data from
# strelka2, mutect2, lancet, and vardict.
plot_oncoplot(maf_list$combined, "combined_maf_oncoprint.png")

# Plot oncoprint using Strelka2 output data and save
plot_oncoplot(maf_list$strelka2, "strelka2_oncoprint.png")

# Plot oncoprint using Mutect2 output data and save
plot_oncoplot(maf_list$mutect2, "mutect2_oncoprint.png")

# Plot oncoprint using Lancet output data and save
plot_oncoplot(maf_list$lancet, "lancet_oncoprint.png")

# Plot oncoprint using Vardict output data and save
plot_oncoplot(maf_list$vardict, "vardict_oncoprint.png")

Session Info

sessionInfo()